« All Episodes

When Sense-making and Statistics Collide

Published 2/24/2022

Our ability to make sense out of things is powerful. When we hear "90% of the way there", our brain fills in the gaps and often stretches that to mean "basically 100%." However, this is a cognitive error. In this episode, we talk about the importance of combining statistical thinking with sensemaking.

🙏 Today's Episode is Brought To you by: Doppler

Thousands of companies of all sizes—from startups to enterprises—use Doppler to keep their secrets and app configuration in sync across devices, environments, and team members. Goodbye .env files.

Get started in under 4 minutes at https://doppler.com/l/developertea

📮 Ask a Question

If you enjoyed this episode and would like me to discuss a question that you have on the show, drop it over at: developertea.com.

📮 Join the Discord

If you want to be a part of a supportive community of engineers (non-engineers welcome!) working to improve their lives and careers, join us on the Developer Tea Discord community by visiting https://developertea.com/discord today!

🧡 Leave a Review

If you're enjoying the show and want to support the content head over to iTunes and leave a review! It helps other developers discover the show and keep us focused on what matters to you.

Transcript (Generated by OpenAI Whisper)
Often in our work, we have to make sense of what we are seeing. And making sense is something that we do quite well as humans, but sometimes in the process of the sense making, we cut out a lot of useful and sometimes even critical information. My name is Jonathan Cutrell. You're listening to Developer Tea, my goal on this show is to help driven developers like you find clarity, perspective and purpose in their careers. I want to do a little experiment with you. Let's imagine that you are tasked with coming up with a list of requirements for a job position. Now let's imagine that as a part of this list of requirements, you also need to come up with some kind of threshold that the prospective candidate needs to meet. In other words, some percentage. If you're looking at your list of requirements, that they need to meet some percentage of those requirements. Now, let's imagine that some of the requirements are kind of considered core requirements. So we'll set those aside and just look at the remainder. One is the threshold that the candidate should meet in order to be considered a good candidate. This is somewhat arbitrary, so we'll go with something like 90%. Our gut tells us that if we find a candidate that could meet 90% of what we want, we could probably make up the rest of that 10%, the missing 10%. Right now, we're experiencing the sense-making part of our brains. We're experiencing the feeling of meeting a candidate and assigning an arbitrary value of 90% good to that candidate. Someone who is 90% good, that sounds like a success to us. This is the kind of narrative part of our brain. We imagine that the number means something and that meaning is what we focus on. But what if we were to look at it from a different angle? Would the story change? That's what we're going to talk about right after we talk about today's sponsor. Developer Tee is proudly supported by Doppler. The scary days of configuring, managing, and sharing secrets across different teams and clouds are over, introducing Doppler, the first universal secrets platform that enables Developer To automate the pain away of managing secrets and environment files. Doppler is your team's central source of truth for secrets and app config across all environments and clouds. Whether your secrets are in Docker, AWS, for cell, serverless, or anywhere else, Doppler supports wherever you're working. And as your stack evolves, Doppler remains simple. More than 11,000 customers across all company sizes from start-up to enterprises use Doppler to keep their secrets and app configuration in sync across devices, environments, and team members. Say goodbye to those dot-in files and set up Doppler for your team in less than four minutes. Head over to Doppler.com slash L slash Developer Tea. That's Doppler, d-o-p-p-l-e-r.com slash the letter L slash Developer Tea. Thanks again to Doppler for sponsoring today's episode of Developer Tee. Let's take our list of requirements. The list that we kind of whittled down, and this, again, is hypothetical. I can't imagine that you would actually write down a list while you're listening to this podcast. But let's imagine that we had, let's say, 20 or 30 items on a list. And these might be specific skills or perhaps traits or experience, specific kinds of experience that you want in a candidate. For those things are, let's imagine that we found a candidate who has met 90 percent of that. Well, specifically, what would that look like? If we play out the statistics directly, rather than trying to derive meaning from that statistic, and if instead we say, let's remove 10 percent of the items on this list, and take the opportunity to randomly select one out of every 10 items and remove it entirely. Another version of this might be that on every item that you see on this list, the kind of completion metric, whatever you would consider, the completion metric has only reached to an average of 90 percent for each of those items. This creates an entirely different feeling from the feeling that we had about having a candidate who had 90 percent of what we want. Interestingly, the imagination that we have when we haven't applied those statistics is very different from the actual picture, the feeling that we get from the actual picture once those statistics have been applied. And the reason for this is because our brains, when we're in sense making mode, we fill in the gaps. In some ways, we imagine that 90 percent is essentially 100 percent. We stretch our imagination to believe that answers are complete answers, not partial answers. If I were to ask you to estimate the number of days that you or even a team member was going to be on your A game, how often are you going to be firing on all cylinders, doing your best at your job, full of energy, all of the things that would be required for you to operate at your kind of optimal capacity. If I were to ask you how many days are you in that state, you would probably have an answer that is lower than a 100 percent. We all have bad days. Even if your day isn't bad, it may not necessarily be that 100 percent optimal day. If you were to say 70 percent of the time, that would probably be a high estimate even for days that you would consider yourself an optimal. Just imagine then that if you combine all the statistics together, that your kind of performance relative to what it would be if you were on top of everything all of the time is actually somewhere around 80 to 85 percent. This doesn't sound unreasonable, but if you were to apply this to your capacity planning or your personal estimates for work that you're doing, you may find yourself feeling surprised. The reason for this once again is because when we're talking about things from a perspective of what's reasonable to expect out of yourself, it's easy to describe those statistics void of the actual work. When we try to apply the statistics to some expected story, some expected narrative, for example, the kind of work or the amount of work that you're going to be able to get done in a given week or month. If you were to imagine that story, it's much easier to imagine feeling good and being productive 100 percent of the time rather than accepting a more realistic perspective, which shows that 80 to 85 percent. One of the reason for this is probably because we don't necessarily like that story. We don't necessarily want to adopt that narrative. We want to see ourselves being successful and full of energy and operating as optimally as possible. But if we want to be accurate, if we want to imagine a more clarified picture of our reality, then we need to find a way to bridge the gap between these statistically more accurate representations of our reality and our narratives. When this is especially pronounced is when we review something that has happened in the past. It's easy to apply statistics and understand something in retrospect. It's much harder to apply statistics in order to predict something. We use sense-making to predict and, generally speaking, statistics to understand. When we're looking retrospectively, it's easy to describe something because all of the pieces fit in place. We can kind of see how things went and none of it seems out of place. But looking forward, it's very hard to predict. And so one of those tools that we can use alongside our sense-making to make our sense more accurate is those applied statistics try to actually carry those statistics out through the picture that you're creating with that sense-making process. Thanks so much for listening to today's episode of Developer Tea. Thank you again to today's sponsor, Doppler. Doppler is your final scalable solution for dealing with environment variables and all of the secrets that you don't want leaking out through an accidental get commit. Go and check it out. Head over to Doppler.com slash L slash Developer Tea. That's d-o-p-p-l-e-r.com slash the letter L slash Developer Tea. Thanks again, Doppler. Thanks so much for listening to this episode of Developer Tea. If you enjoyed this discussion, now I encourage you to join our Discord community. It's 100% free. Come in and ask a question. This is the most important thing you can do to get a conversation rolling. Come in with the questions that you have. Even if it seems like it's a completely off-the-wall question, it's going to bring something to the group and we need you to be a part of that. Go and check it out. Head over to developertea.com slash Discord. Thanks so much for listening. And until next time, enjoy your tea.